Anthrax lethal toxin induces cell death-independent permeability in zebrafish vasculature

Bolcome et al. 10.1073/pnas.0712195105.

Supporting Information

Files in this Data Supplement:

SI Figure 5
SI Movie 1
SI Figure 6
SI Figure 7
SI Figure 8
SI Table 1
SI Materials and Methods

Fig. 5. Zebrafish CMG2 orthologues closely resemble human CMG2. Alignment of human ANTXR2/CMG2 with predicted sequences of the zebrafish CMG2A and CMG2B genes. Regions of identity are highlighted in yellow. Protein domains were predicted by SMART (1). The putative signal peptide region, von Willebrand A (vWA) domain, and transmembrane domain are indicated in orange, blue, and gray text respectively. Residues of the metal ion dependent adhesion site (MIDAS) motif are highlighted with green. Conserved Tyr-119 and His-121 residues, which are important to PA-CMG2 binding, are highlighted in blue (2). Cytoplasmic prolines and tyrosines that may be important in receptor signaling are highlighted in red. RT-PCR was used to confirm that CMG2A and -B are both expressed in zebrafish embryos before and at the time of injection (data not shown).

1. Schultz J, Milpetz F, Bork P, Ponting CP (1998) SMART, a simple modular architecture research tool: identification of signaling domains. Proc Natl Acad Sci USA 95:5857-5864.

2. Santelli E, Bankston LA, Leppla SH, Liddington RC (2004) Crystal structure of a complex between anthrax toxin and its host cell receptor. Nature 430:905-908.

SI Movie 1

Movie 1. Zebrafish embryos injected with LeTx display trapping of blood cells within the heart and reduced blood flow at late stages. The Tg(gata1:dsRED) (1) transgenic line was used. First segment (Dye injected control 20 hpi) shows a WT embryo at 72 hpf, with blood cells flowing through the heart and vasculature. In the second segment, an embryo with mild LeTx phenotype is shown with blood cell accumulation in the heart and reduced circulation. In the final segment, an embryo displaying severe LeTx phenotype is shown where blood cells are trapped in the heart without blood flow.

1. Traver D, et al. (2003) Transplantation and in vivo imaging of multilineage engraftment in zebrafish bloodless mutants. Nat Immunol 4:1238-1246.

SI Figure 6

Fig. 6. Progression of LeTx phenotype over time. (A) LeTx phenotypic stages 1-4 are depicted in each row. Stage 1: Left panel shows the beginning of inflow tract regurgitation (arrow) as compared to embryos injected singly with either PA or LF (Center and Right panels, respectively). Stage 2: Left panel shows increased blood accumulation in the heart chambers. Right panels show the progressive collapse of an ISV lumen at this stage. Fluorescent panels show the progressive decrease in lumen size of the same endothelial cell over three time points after LeTx injection. Most ISVs lose their function by the end of stage 2. Stage 3: Pericardial edema becomes pronounced, and the lumen size of the common cardinal vein (CCV) is highly reduced at the end of stage 3. Fluorescent images depict CCV lumen reduction at two time points after LeTx injection. Stage 4: In embryos with a severe phenotype, the outflow tract of the heart closes completely, preventing erythrocytes from exiting the heart from the ventricle. Left panel, arrow in bright field image denotes remaining pooled blood in the inflow tract from stage 3 that will enter the heart chambers by the end of this stage. Transgenic zebrafish expressing red fluorescent blood cells Tg(gata1:dsRED) (21), were used to compare the outflow tract between dye (Center) and LeTx injected (Right panel). Hour post injection time discrepancies in fluorescence images are due to variability in the onset time of toxin phenotypes. A, atrium; V, ventricle. (Scale bars, 40 mm.) (B and C) Individual red blood cells flowing through the ISVs anterior to the cloacae (B), or the subintestinal vein (C) were counted for control or LeTx injected Tg(gata1:dsRED) embryos, for three incremental doses of LeTx at 20 hpi. In mild embryos, there were statistically significant differences in the numbers of blood cells flowing through each region between embryos receiving 1´ and 4´ LeTx doses (P < 0.001 for both regions) as well as between those receiving 2´ and 4´ doses (P = 0.018 and 0.003 for the intersegmental arteries and SIV respectively). A 1´ LeTx dose is 37 fmol of LF and 25 fmol of PA. Statistics were completed using the Holm-Sidak method. (D) The zebrafish embryonic heart rate was not altered by LeTx exposure at 20 hpi (P < 0.001). Error bars denote standard deviation.

Fig. 7. LF_NDTA and PA produce a different phenotype from LeTx. Zebrafish embryos were injected with 25 fmol of PA plus 11 fmol of LF_NDTA (A) or LF_NDTA alone (B) at 48 hpf and photographed at 20 hpi as in LeTx experiments. (Scale bar, 275 mm.) (C--F) Nuclei counts using the TG(fli1:nEGFP)^y7 (1) line were used to determine endothelial cell numbers over the 8 ISVs anterior to the cloacae between dye injected controls (C and E) and embryos injected with LF_NDTA plus PA (D and F). Endothelial nuclei numbers began declining in most embryos by 6 hpi and were almost entirely gone by 20 hpi in all toxin injected embryos (N = 2, n > 20; P < 0.001). A representative LF_NDTA plus PA-injected embryo is depicted at 6 hpi (D). (Scale bar, 80 mm.) (G and H) LF_NDTA and/or PA effects on ISVs labeled by EGFP are as indicated. (I) Cycloheximide was used at a concentration of 5 mM for 6 h to phenocopy the effects of translational inhibition. (Scale bar, 80 mm.)

1. Roman BL, et al. (2002) Disruption of acvrl1 increases endothelial cell number in zebrafish cranial vessels. Development (Cambridge, UK) 129:3009-3019.

Fig. 8. Inhibitor dose curves for ZM323881, SU11652, and PTK/ZK. Each inhibitor attenuated LeTx effects in a dose-dependant manner. Effective inhibitor doses were observed from nanomolar to low micromolar concentrations. A 2´ LeTx dose was used (75 fmol of LF and 50 fmol of PA). Based on these data, the most effective doses for each inhibitor were: 1.5 mM SU11654, 1 mM ZM323881, and 500 nM PTK/ZK, which were chosen for attenuation of LeTx effects in subsequent experiments (SI Table 1).

SI Table 1

SI Materials and Methods

Isolation and Characterization of Zebrafish ANTXR cDNA Clones. The human amino acid sequence for CMG2 was used to BLAST search the GenBank zebrafish EST library for translational matches. A 5' primer (5'-GCGGATCCGCCGTCATGACAAAGGAAAATCTCTGG-3') for the CMG2A transcript was designed from GenBank Accession No. CA471164. A 3' primer (5'-CGGAATTCTCACAGATCCTCTTCTGAGATGAGTTTTTGTTCATGCTGCGTGCGACTG-3') for CMG2A incorporating a myc-tag was designed from GenBank Accession No. BI475178. A 5' (CCTCTAGAGCCACCATGAGAGGAGACAGCA) and 3' (CGAATTCGAAGCCCTTATCATTTGCTGTACC) primers for CMG2B were designed using GenBank Accession No. XP_689332.1 and No. XM_684240.1, as well as Ensembl gene ENSDARG00000063011. RT-PCR was used to clone CMG2A and CMG2B into pCR II-TOPO vectors (Invitrogen) from 48 hpf embryos. GenBank accession numbers are DQ415957 and EF591979 respectively. We also identified duplicated genes and isolated partial cDNAs for zebrafish ANTXR1A/TEM8A and ANTXR1B/TEM8B (R.E.B. and J.C., unpublished results).